Instructions to use tiiuae/Falcon-H1-Tiny-R-90M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiiuae/Falcon-H1-Tiny-R-90M with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="tiiuae/Falcon-H1-Tiny-R-90M")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tiiuae/Falcon-H1-Tiny-R-90M")
model = AutoModelForCausalLM.from_pretrained("tiiuae/Falcon-H1-Tiny-R-90M")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tiiuae/Falcon-H1-Tiny-R-90M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiiuae/Falcon-H1-Tiny-R-90M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiiuae/Falcon-H1-Tiny-R-90M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tiiuae/Falcon-H1-Tiny-R-90M

SGLang

How to use tiiuae/Falcon-H1-Tiny-R-90M with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "tiiuae/Falcon-H1-Tiny-R-90M" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiiuae/Falcon-H1-Tiny-R-90M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "tiiuae/Falcon-H1-Tiny-R-90M" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiiuae/Falcon-H1-Tiny-R-90M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tiiuae/Falcon-H1-Tiny-R-90M with Docker Model Runner:
```
docker model run hf.co/tiiuae/Falcon-H1-Tiny-R-90M
```

ybelkada commited on Jan 13

Commit

a770de8

verified ·

1 Parent(s): a5dc8ab

Upload FalconH1ForCausalLM

Browse files

Files changed (3) hide show

config.json +68 -0
generation_config.json +10 -0
model.safetensors +3 -0

config.json ADDED Viewed

	@@ -0,0 +1,68 @@

+{
+  "architectures": [
+    "FalconH1ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attention_in_multiplier": 1.0,
+  "attention_out_multiplier": 1.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "embedding_multiplier": 0.034423828125,
+  "eos_token_id": 11,
+  "expansion_factor": 1.5,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 512,
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
+  "key_multiplier": 1.0,
+  "lm_head_multiplier": 0.078125,
+  "mamba_chunk_size": 128,
+  "mamba_conv_bias": true,
+  "mamba_d_conv": 4,
+  "mamba_d_head": 32,
+  "mamba_d_ssm": 768,
+  "mamba_d_state": 64,
+  "mamba_expand": 2,
+  "mamba_n_groups": 1,
+  "mamba_n_heads": 24,
+  "mamba_norm_before_gate": false,
+  "mamba_proj_bias": false,
+  "mamba_rms_norm": false,
+  "mamba_use_mlp": true,
+  "max_position_embeddings": 262144,
+  "mlp_bias": false,
+  "mlp_multipliers": [
+    1.0,
+    1.0
+  ],
+  "model_type": "falcon_h1",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "num_logits_to_keep": 1,
+  "pad_token_id": 0,
+  "projectors_bias": false,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 100000000000.0,
+  "sliding_window": null,
+  "ssm_in_multiplier": 1.0,
+  "ssm_multipliers": [
+    1.0,
+    1.0,
+    1.0,
+    1.0,
+    1.0
+  ],
+  "ssm_out_multiplier": 1.0,
+  "tie_word_embeddings": true,
+  "time_step_floor": 0.0001,
+  "time_step_max": 0.1,
+  "time_step_min": 0.001,
+  "time_step_rank": "auto",
+  "transformers_version": "4.57.0",
+  "use_cache": true,
+  "vocab_size": 32768
+}

generation_config.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    11,
+    228
+  ],
+  "pad_token_id": 0,
+  "transformers_version": "4.57.0"
+}

model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:1556cfc66e5a4b49cf1058c1fb02f3ffcfb191304bf9021066d6e0d05f473a44
+size 182304120